"Un-default" Behavior Tuning: Specifying Model Behavior outside the Norm with LLM Self-Playing and Self-Improving
Abstract
Specifying model behavior is challenging—especially when the desired behavior is unpopular relative to the model’s training data. Reversing the influence of massive training corpora is both time-consuming and costly, and such interventions are typically inaccessible to end users. While Large Language Models (LLMs) make it easier to write instructions using natural language, specifying unpopular behaviors remains a difficult task.
Fields of Study & Subjects
Multi-Agent Systems and Negotiation
Embodied and Extended Cognition
Explainable Artificial Intelligence (XAI)
Publication details
- Journal / source
- Proceedings of the 31st International Conference on Intelligent User Interfaces
- Publisher
- ACM
- Pages
- 699–716
- Publication date
- 3 March 2026